On 31 March 2020 NSE had about 3.1 crore unique registered investors. On 31 August 2026 it had 13.5 crore. More than 10 crore Indians opened their first trading account in six and a half years, most of them after watching someone they knew make money. That is the most powerful herd in Indian market history, and mostly it has been a good thing. Long term savings moved into equity. It also means a very large number of investors learned the market from the crowd, at the speed of a group chat.
Herding is not stupidity. Following others is often a sensible shortcut. The trouble starts when everyone is following everyone, and nobody is looking at the business any more. I have watched that switch flip more than once on trading desks. It is always obvious afterwards and rarely obvious at the time.
unique registered investors on NSE, 31 Aug 2026, up from 3.1 crore in Mar 2020
NSE Market Pulse, Sep 2026
of investors use friends, family and colleagues as an information source
SEBI Investor Survey 2025
of investors who follow finfluencers make some or most decisions on their advice
SEBI Investor Survey 2025
Why herding is individually rational but collectively catastrophic
Herd behaviour means following other people's decisions instead of forming your own. Three forces make it feel rational.
- Information cascades: other people's actions carry information. If thousands are buying, perhaps they know something. Bikhchandani, Hirshleifer and Welch showed in 1992 how, after a few early movers, it becomes rational for each new person to copy the crowd and ignore their own information. The chain then carries no new information at all, only the first few guesses, amplified.
- Career risk: a professional who underperforms with the same stocks as everyone else keeps the job. One who underperforms with different stocks may not. So portfolios drift towards the benchmark and towards each other.
- Social proof: when people around you are making money in a theme, staying out feels like being wrong in public. The pressure is social before it is financial.
Each of these is reasonable for one person on one day. Added up across millions, they push prices away from value on the way up and through it on the way down.
SEBI, Investor Survey 2025, released 30 Sep 2025Where the Indian herd gets its information
SEBI's Investor Survey 2025 covered more than 90,000 households. Among investors, the top sources of information were friends, family and colleagues (59%), financial influencers on social media (56%) and online investment groups on messaging apps and forums (34%). Investor education by institutions such as SEBI, NISM, the exchanges and AMFI reached 15%. Of those who follow finfluencers, 93% found them at least moderately credible, and 62% made some or most of their decisions based on their recommendations.
That is a cascade waiting to happen. When the main source of information is other investors, the same idea travels fast and lands on everyone at once. SEBI has acted on the supply side. In 2024 it barred regulated entities from associating with unregistered persons who give investment advice or make return claims. The demand side, the urge to do what everyone else is doing, no rule can remove.
Unique registered investors on NSE, lakh, at the end of each financial year and on 31 Aug 2026.
Unique registered investors on NSE: 310 lakh in FY20, 400 in FY21, 594 in FY22, 727 in FY23, 916 in FY24, 1,128 in FY25, 1,291 in FY26 and 1,352 lakh on 31 Aug 2026.
NSE Market Pulse, September 2026 (unique registered PANs). Investors under 30 were 37.7% of the registered base in Aug 2026, against 23.5% in Mar 2020.
How bubbles form and burst, the Indian pattern
Every cycle has a different theme. The mechanics repeat. Economists from Hyman Minsky to Robert Shiller have described roughly the same sequence.
- A real storyA theme with genuine fundamentals attracts informed money. Price follows earnings.
- Momentum moneyRising prices attract traders who buy because prices are rising. Valuations stretch.
- The crowd arrivesNew investors see others making money. Account openings and inflows jump. Social media amplifies.
- This time is differentEvery concern is dismissed. Volumes peak. Price now leads the story.
- The unwindEarly money exits, momentum money follows, the crowd holds the top.
India's own record, using NSE index closes:
- Nifty 50 falls 59.9%
From 6,287.85 on 8 Jan 2008 to 2,524.20 on 27 Oct 2008, after a boom led by real estate and infrastructure stocks.
- Nifty 50 falls 38.4% in ten weeks
From 12,362.30 on 14 Jan 2020 to 7,610.25 on 23 Mar 2020. Not a bubble, a shock, but the panic selling looked the same.
- 193.4 lakh new investors
The largest wave of first time investors, straight after the 71% rally of FY21.
- Nifty Smallcap 250 down 26.1%
From its record 18,623.15 (P/E 32.15) on 23 Sep 2024 to 13,755.90 (P/E 25.92) on 3 Mar 2025.
- Smallcap 250 back within 0.5% of its record
18,527.20 on 8 Sep 2026, at a P/E of 34.19. It had still not closed above the record on 12 Oct 2026.
The 2024 to 2025 small cap episode is the cleanest recent lesson. At its September 2024 record the Nifty Smallcap 250 traded at a trailing P/E of 32.15. Five months later it was 26.1% lower at a P/E of 25.92. Then it climbed 34.7% off the low, to within 0.5% of the record by September 2026, which is the part herd sellers missed. Herding hurts twice: once on the way in at the top, and again on the way out near the bottom.
The SME platform saw its own version. NSE's SME segment has hosted 748 IPOs since inception, raising about ₹23,993 crore by July 2026. After a period of heavy oversubscription in 2024, SEBI tightened SME IPO eligibility in March 2025, including a minimum operating profit of ₹1 crore in two of the previous three years and a cap on the offer for sale at 20% of the issue. When a corner of the market needs new rules to cool demand, the crowd has usually arrived.
What would you do?
It is 3 Mar 2025. The Nifty Smallcap 250 is 26.1% below its September 2024 record. Your group chat is full of people saying they have sold.
- Sell with the group
You lock in the fall at 13,755.90. The index closed at 18,527.20 on 8 Sep 2026, 34.7% higher.
- Follow your written plan, for example keep the SIP and rebalance on the date you set
Your instalments in the following months bought at 13,844 to 15,360, the cheapest prices of the cycle.
- Double your small cap exposure because it is cheap
It worked this time, but the P/E was still 25.92 and the index fell again in Mar 2026. Size decided in the moment is the herd in reverse.
NSE closes and P/E for the Nifty Smallcap 250, Sep 2024 to Sep 2026. Past prices, not a forecast.
FOMO, the herd behaviour amplifier
FOMO, the fear of missing out, is the emotion that pulls people into a theme near its peak. When a stock or sector has already doubled and your feed is full of screenshots, staying out feels unbearable, even when you know the price is stretched.
The trap is an asymmetry. After a big run the remaining upside is usually smaller and the downside larger. FOMO makes it feel the other way round. The SEBI survey picked this up directly: among people who intend to start investing, 23% named "quick gains with small investments" as a trigger.
MythIf everyone I know is buying it, the risk must be low.
FactCrowd buying raises the price and the valuation, which raises the risk. The Smallcap 250 traded at a P/E of 32.15 at its Sep 2024 record, and above 33 a few weeks later, just before a 26.1% fall.
Common mistakes
Mistake: treating a crowded trade as a confirmed one. If you heard about it from a group chat, an influencer and a relative in the same week, you are probably not early.
- Copying the entry and not the exit. The person who told you to buy will not message you when they sell.
- Selling with the crowd at the bottom. Herding is just as costly on the way out.
- Confusing a fast rise with a strong business. Check earnings and valuation, not how many people are talking about it.
- Following unregistered advice. Check whether the person is SEBI registered before acting on any recommendation.
RupeeCase index algos buy and sell on quantitative factor signals, not social sentiment. A stock that has risen sharply can be dropped at the next rebalance if its momentum fades or its quality score slips, and a stock nobody is talking about can be added if it qualifies. Momentum rules do ride crowds, but they enter and exit by rule, on a fixed date. See the strategies and every past list on the rebalance history page. Figures there are model, back tested.
Questions readers ask
Is momentum investing just herding with a spreadsheet?
Partly, and honestly so. Momentum strategies profit from the fact that crowds move prices gradually. The difference is the exit: a rule sells when the trend fades, on a fixed schedule, while a herd usually sells after the fall.
Can you spot a bubble in advance?
You can spot conditions: valuations far above their own history, a surge of first time investors, and a story that dismisses every risk. Timing the top is much harder. That is why position limits and rebalancing work better than predictions.
How do I check whether someone giving tips is registered?
SEBI publishes searchable lists of registered investment advisers and research analysts on its website. If a person recommends specific stocks for a fee or promises returns and is not on those lists, treat the advice with great caution.